Topic: Bitcoin Scalability? (Read 1968 times)

Risk of what

One of the great features of Bitcoin is that it removes counterparty risk.
By using Lightning transactions you are essentially introducing a counterparty risk if the payment hub provider
has nefarious intentions. Therefore really high value transactions will probably still be settled on the main
blockchain even if Lightning is adopted by a huge part of the Bitcoin user base.

Disclaimer:
I´m just playing devil´s advocate here, actually I´m fairly optimistic about
the potential of Lightning and other 2nd layer solutions.

What are your security worries? You seem to be afraid of Lightning transactions, but not about anything specific.

If you do not settle on-chain where is then the bitcoin security ?  I can use PayPal here - so you sell vaporware and have no disclaimer in a financial env.

No bigger company with proper due diligence buys this .

Nobody at all should buy this - sorry.

Hi guys,

this is our idea to scale Ethereum. But you may borrow it for Bitcoin, as it would work for any blockchain:

http://afterether.org/blockchain-scalability-by-blockchain-clustering.html

would be very interesting to know what do you think of it

This is not true.

Lightning does not need to be settled on-chain. This is wrong. Once money is on the Lightning network, it can stay there until a user wants to use it on-chain instead. Settling on-chain is not necessary at all, and so because mining is not involved in Lightning transactions, fees can be cheap or free (the only cost is running your computer and it's Bitcoin node, basically nothing).

For instance, if your friend sends some money directly to you with Lightning, you can charge a fee if you like. Just don't expect them to still be your friend if you do (or to receive the money soon )


This will not require 100 MB blocks or anything like that, as there can be huge amounts of money on Lightning all the time. Lightning transactions don't use block space, they are off-chain.

Yes - all those 2nd layer techs need to settle on-chain and will either need more block-size or higher fees. Making ordinary on-chain tx way to costly for small users ( yet told to run a relay node each).

The biggest issue with these new techs is, that they are brand new compared to on-chain transactions and so not really same secure or rather insecure by many orders of magnitude.

All ppl that promote those things will mostly not tell you about risks and disclaimers as it would be usual for any financial solutions.

Few hundred transactions per second? Not for current nodes in my opinion.
Look at this; https://blockchain.info/unconfirmed-transactions > there are over 30,770 unconfirmed transactions.
Total Fees      5.03496607 BTC
Total Size        49017.571 (KB)
Transactions Per Second   3.88

Every block included into blockchain every 10 minutes, and most miners submit about 1Mb for each block.
There is 49 Mb unconfirmed transactions in waiting list right now.

This is a pretty solid reason to not go for the block-size increase. Is there somewhere one can find proof to this. Like the work involved in syncing blocks and time a typical computer would take for that.
This way, we can know what hardware acceleration will support or be necessary to allow and increase in the block-size without these issues.

Woldn't implementing lightning network require even bigger blocks at some point? Like 100 MB block size to cover all the needs of network for billions of people?

This might become a non-issue soon. The Lightning Network will take all these micro transactions clogging up the network now and replace it with side channels. These side channels will handle the load and only the balance will be pushed to the Blockchain with one transaction.

The Block size can stay small, because the load is handled by side channels on the Lightning Network. This will prevent a massive Blockchain and reduce bandwidth requirements for running a node, making it more viable for people to run full nodes and to increase the decentralization of the network.

This does not mean that the Block size will stay the same. If the single tx's coming back from the side channels increase to a level where the Block size needs to increase, then it can still be done. ^smile^ < This will just happen less regularly, because most tx's are handled by the Lightning Network >

Will bitcoin every be able to handle more than a few hundred transactions per second ?

I have to correct myself.

The actual quantity of a token used as a medium of exchange does not matter; e.g. the total supply of physical gold is never too much or little for it to act as money; scarcity is only relative, never absolute. We may choose whatever small amount of gold to represent a unit of currency - just don't choose an amount such that a cup of coffee costs EUR 2,000,000! Even the satoshi could further be divided into a million parts when needed through an implementation change. 
     
One of the quality needed for a token to be money is that the total supply is evenly distributed among the population. Even though Bill Gates's wealth is into the billions, his personal holding of the cash USD is only an miniscule  part of the total US money supply. But Bitcoin is not distributed throughout the world's population and so can never serve as money.

Bitcoin may serve as a store of value just as the precious metal or other valuables. But currently, its scarcity viz-a-viz the fiat makes it only as a tool of speculation. On average in the longer term, any holder of bitcoins among the lower 95% would have got them at the "wrong" price making it a negative investment for the future.       

You could use mBTC, microBTC or even satoshi as "everyday currency", so the essential problem is not how valuable one BTC is. However, there will be a problem when satoshi is more than, let's say, 1 cent. Then you have no way to deal with small purchases.

I partially agree with this in that there may be a better solution in the future however, I disagree with the fact that the problem is '1 desktop = 1 vote'; The real problem is finding a way to put the power of authority into the masses (i.e. decentralisation). So far, there hasn't been a single cryptocurrency that is truly decentralised - POW = centralised mining power. POS = centralised power to largest stakeholders.

The closest thing that we have to decentralisation is The Tangle (IOTA) or Delegated proof of stake (DPOS). Although, both protocols still have their downfalls and disadvantages that stop them from either, working efficiently or solving the economic problem (scarcity*). DPOS also still puts the power into the few who are successfully elected and voted for.

*When I refer to scarcity, I refer to the economic problem NOT scarcity in itself.

It is true that my knowledge of how the bitcoin network works is rudimentary. My "attack" on bitcoin is in order to have replies to clarify doubts that I have gathered about bitcoin. I think most of those posting in "Bitcoin Discussion" don't really have a clue how the blockchain works.
I have this conclusion about bitcoin as it is currently:
1) Bitcoin will never and can never replace fiat.
Hosts of problems: the 21 million limit means super scarcity as compared to national fiat. Singapore's M2 2017 at SGD 600,000 million; means rate of 600,000/20 = SGD 30,000 per 1 BTC. What about with USD.  BTC may at most be IMF "reserve currency" as clearance currency between central banks; how does such a status be consistent with a medium of exchange.
2) Bitcoin may only be a medium of speculation.
Bitcoin has no practical use. Even if the proposed off-chain Lightning Network works, the natural volatility of bitcoin will never make any system of bitcoin payment feasible.
   By definition, the world of speculation means the bottom 95% of small investor will always suffer lost. In the current rally to above USD7500, these 95% will never ultimately be sitting happily with their "profits" riding the wave upwards. By definition, the correction would be a bloody for the 95%.     
3) Distributed consensus is incompatible with mining hardware advantage.
The current situation with hash power concentrated in the hands of a few miners means this Bitcoin network will not be of any use for our economic life. Crypto will only start to have real benefits to human economic activity only when someone comes up with a true innovation where simple desktop acts as a "vote". Yes! This is the real challenge! Where ordinary users have monetary incentives to participate in the network and no one with money advantage can build "mining rigs" to monopolize the  network. This is the only true open source.   

People seem to be fixated to this method of "proof-of-work". In the near future, I believe there will be ideas now totally unexplored that may be far far better than our current over-hyped blockchain technology. Don't underestimate true human ingenuity .   

Truth be told, Bitcoin is scalable. It will never reach a threshold, blocksize can be increased to fit more transactions, better internet, and cheaper disk storage. This though is limited by time, if you are ready to wait more 5 years allow 1 million transactions, then it would scale. If you want to do it today, it cannot.

Hence, we try to build ways around it. Lighting Network, and rest are just merely tweaks to make it scalable and are working just fine. If you can compress the blockchain without pruning the blockchain, then it would scale to infinite transactions today itself. Why compress ? because if you put more and more transactions the size of block increases, since this is a decentralized system and the block has to travel to all full nodes with varying internet connection, puts constrains on the system.

To scale blockchain today, without sacrificing its important qualities such as decentralization and immutability is a hard problem.

The natural assumption is that CPU mining would be more distributed, but it's by no means a guarantee.  Before GPU mining and ASICs became a thing, Bitcoin used to have a bit of an issue with Botnets.  One attacker with an exploit could infect thousands, or even millions of desktops around the world and have them all mining Bitcoin without the owners' knowledge or consent.  That means unscrupulous actors could gain a larger majority in determining how the network is run.  In any system, there will always be those who attempt to find ways to gain an unfair advantage.

Also, as a small quibble, strictly speaking, nodes don't "vote".

That conspiracy theory is categorically false. A lot of your "questions" are things that are leftovers from Satoshi's designs that require a hard fork to change, and we are hesitant to have hard forks willy nilly.

That is a natural progression of technological advancement. A Proof of Work system will always result in an arms race of who can get the best and fastest mining hardware. It doesn't matter whether it's CPUs, GPUs, or ASICs. Those who have the most money to sink into building massive mining farms are those who are going to have an advantage. Proof of Stake isn't any better (in fact it's worse and has several of its own severe issues).

Why is there no need? We need fees and the block subsidy to incentivize miners to continue to mine and secure the blockchain. Without them, mining would not have had a start at all and there would be no incentive to mine blocks. Remove those incentives now and miners will have no incentive to continue to mine blocks, so mining will stop.

They would, if there were an incentive to do that. There is no incentive to continue to mine (and expend energy and money) if there were no incentives. Furthermore, the difficulty of mining is so high that the combined computational power of people's desktops is likely not enough to mine blocks at the current pace. Not only that, but people also want to be able to use their computers to do other things too, so you can't expect everyone to dedicate all of their computing power to mining. Having the difficulty this high has benefits too; it makes performing attacks much much harder. Having a high difficulty means that the blockchain is very secure; it is extremely expensive and computationally difficult to rewrite history. That is a good thing.

While have "one node one vote" is a great idea in theory, reality tells us that this is basically impossible. Nearly all Proof of Work algorithms have been shown to be performed on GPUs, so in order for "one node one vote", every node would need to have several powerful GPUs to mine. Furthermore, as technology gets better, it is likely that such algorithms will eventually have ASICs built for them; nothing is really ASIC resistant.

And you cannot stop me from telling you that such statements are wrong and very harmful to the community. Furthermore you have provided no evidence that "the developers have been bought", just nonsense that shows you don't understand how this system works.

No they could not. Such a change requires a hard fork and requires all miners and users to upgrade their software to support that. It is highly unlikely that the Bitcoin Core developers would be able to convince the entire Bitcoin community to adopt such a change.

The Bitcoin Core developers have been horrifically bad at "sweet talking" people into doing what they want. Look at how long it took to activate segwit; if the developers could "sweet talk" people into doing what they want, why wasn't segwit activated as soon as possible? For that matter, why haven't all forks activated as soon as they could (CSV did not, CLTV did not, etc.)? The Bitcoin Core developers would not be able to "sweet talk" anybody into doing anything; they're good with code, not words.

Except those central bankers don't decide on Bitcoin or its money supply and you have provided no evidence to back up your claim, just nonsense statements that are clearly false.

The maximum block size was already increased with Segwit to 4 MB. Segwit2x saying that it increases the block size to 2 MB is completely false; it increases the block size to at most 8 MB. To say that it does not is false and lying. The absolute size of blocks currently have already exceeded 1 MB since segwit activated.

The Bitcoin Core developers have already stated why they are opposing Segwit2x. It has been widely published already. I'll give you a quick run down:

• The fork was done extremely hastily with little to no apparent testing or preparation
• The fork does not do anything that is on the hard fork wishlist to fix any of the bugs that currently exist in Bitcoin that require a hard fork to fix. It literally does one thing, but hard forks should include as many things as possible to reduce the need to fork again later
• 8 MB blocks are currently too large for the network to handle. They are still vulnerable to several attacks and makes those attacks worse. 8 MB blocks will increase the requirements for running a node (more computing power, network bandwidth, storage space, etc.) and as such will likely decrease the already fairly low node count.
• We have only recently activated Segwit and we don't know how that will effect Bitcoin and the economy. We should wait for Segwit to be actively used and assess later whether a further block size increase is necessary.

I see the 1MB block size (now, max block-weight under SegWit) as a Schelling point. As soon as you grant the principle that the block size can be changed, then it can be changed again and again, in which case, it is the same as having no block size limit at all. But we already know that having no block size limit is bad for a lot of reasons - it will cause centralization of the network, it will degrade service, it will attract an unlimited amount of lint and cruft into the blockchain, and can even be used to orchestrate DoS attacks on the network. So, the line has to be drawn somewhere. 1MB (that is, max block-weight) is that line.